Byrne Hobart
Longreads
- Gwern has a lengthy meditation on effort, specifically the kind of effort involved in doing what was previously thought to be impossible. It's not so much a unified theory of extreme effort as it is a comprehensive refutation of the existence of such theories. There are stories about people making a discovery because they're told it's possible (AI teams recreating other teams' work), because they aren't told it might not be possible (assorted famous stories of mathematicians seeing problems on the board that they think are homework but turn out to be open problems), and there's at least one case not mentioned where someone launched a project specifically because he was told it was impossible (Ed Thorpe asked Richard Feynman if it was possible to beat roulette, and Feynman said no, so Thorpe did it). The piece closes with some darker thoughts—what if extreme success is the more visible part of some tradeoff that most people aren't willing to take?
- Derek Thompson in The Atlantic on the antisocial century: time spent home alone soared during the pandemic, and mostly hasn't reverted. In fairness to the median American, at-home entertainment keeps getting better. But there are negative costs to this: online life is optimized against serendipity except of a particular kind (the services you use want to show you a little randomness so they get a sense for which tastes you haven't expressed, but they're still doing this by looking at the behaviors of people who are otherwise similar to you). More on the economics of this in yesterday's subscribers-only issue ($).
- Misha Saul interviews Tyler Cowen. They're both opinionated people who are interested in basically everything, so it's a wide-ranging discussion: when you're judging the British Empire, how do you weigh the upside of America against the downside of holding back India's development? How should people of different faiths read the Old Testament? How have internal immigration patterns affected American music? And what's the deal with agency?
- Hannah Ritchie looks at how the famous Simons/Ehrlich bet would have turned out in various decades. This was a famous bet on whether a basket of commodities would rise or decline in price over the course of a decade, as a proxy for whether the human drive to innovate would overcome materials shortages. Though arguably, it was the wrong structure for an interesting question: the better technology gets and the more abundant it is, the more demand we'll have for resources because there's more to do with them, and if we're wealthy and technologically advanced, we'll keep finding ways to extract the next batch.
- Tanner Greer has observations on India. The most interesting claim in the piece is that India is the largest-scale and most successful example of post-liberal politics ("post-liberal" as in rejecting the broad set of ideas downstream from J.S. Mill, not "post-liberal" in the sense that a left-leaning party lost an election). The piece is also a good reminder that there's wide variance in wealth and state capacity among developing countries: the author notes that when he visited India, he needed to take antimalaria pills, and that in China such precautions aren't necessary. This is the kind of thing that differentiates a country with a GDP per capita of $13k from one with a GDP per capita under $3k.
- This week in Capital Gains, we consider whether public sector companies that perform the same tasks as private-sector ones can offer better prices because they don't need to earn a profit. Answer: no; they still have a cost of capital, even if they don't account for it, and one of the private sector's advantages is that returns compared to cost of capital is a strong signal about where future capital should be allocated.
Books
More Than a Numbers Game: A Brief History of Accounting. Accounting is one of those messy disciplines that tries to apply repeatable rigor to complex topics on which opinions will always differ. The usual result of this is that if you look at any given accounting rule, you can immediately start to poke holes in it, but for any of those objections you find, it's very likely that someone considered the alternative you're proposing and learned the hard way not to do it. Even some of the ongoing deficiencies in accounting logic have iternal justifications of their own: most big companies are mostly valued based on intangible assets, but those assets are simply harder to value than the tangible ones, and if accountants were unusually good at assessing the true economic value of a brand name or network effect, they'd presumably monetize that skill in asset management instead.
This book traces the history of modern corporate financial statements through two ancestors: first, the statements that overseas lenders wanted from the American railroads whose bonds they bought, and second, the internal reports that various manufacturers used to figure out whether the various product lines they worked on were making any money. These are different questions! A bond investor obviously cares about downside, and if they're providing capital for some new investment, protecting that downside consists mostly of ensuring that valuable assets were purchased with their funds. Upside matters, but as a second-order consideration—if you're a buy-and-hold bond investor who's buying safe securities, you only start thinking about the cash flow required to pay interest if either the underlying assets are impaired or the cash flow isn't there. But companies toggling between different product lines have a different problem: they're maximizing upside rather than controlling downside. They almost invert this formula: if a car company is spending a lot of time asking about the value of inventory and factory equipment, it's because the business of selling cars looks less attractive than the prospect of liquidation.
Equity owners straddle these two groups. They're interested in downside, but also in upside. And they can essentially borrow from both schools of thought to get the information they want. What this book does exceptionally well is to tell the history of continuing compromise and iteration: corporate disclosure has evolved from just disclosing dividends to offering comprehensive balance sheets, income statements, and a cash flow statement that highlights how these two interrelate, along with extensive notes on the assumptions used to prepare them. And for all the complaints about general accepted accounting principles' limitations and the puffery of some non-GAAP numbers, the system works surprisingly well.
Open Thread
- Drop in any links or comments of interest to Diff readers.
- What are some other good histories of evolving norms like More Than a Numbers Game? Is there something similar for law, medicine, engineering, etc.?
Diff Jobs
Companies in the Diff network are actively looking for talent. See a sampling of current open roles below:
- A growing pod at a multi-manager platform is looking for new quantitative researchers, no prior finance experience necessary, 250k+ (NYC)
- Ex-Ramp founder and team seek a high energy, early career full-stack engineer to help build the automation layer for the US healthcare payor-provider eco-system.(NYC)
- A premier proprietary trading firm is looking for smart generalists to join their investor relations team, working with external investors, rating agencies, and the internal finance team. Investment banking and/or investor relations experience preferred. Quantitative background and technical aptitude a plus. (NYC)
- YC-backed AI company that’s turning body cam footage into complete police reports is looking for a tech lead/CTO who can build scalable backend systems and maintain best practices for the engineering org. (SF)
- A well-funded startup that’s building the universal electronic cash system by taking stablecoin adoption from edge cases to the mainstream is looking for a senior full-stack engineer. Experience with information dense front-ends is a strong plus. (Singapore)
Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.
If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.